Motion Verbs Completion Study

This experiment was a control study which tested the telicity of verbs from the self-paced reading study.  Method  For the paper-and-pencil sentence completion study, the 32 experimental sentences from the reading time experiment were used. To prompt completions, the sentences were truncated after the verb. Note that the sentence fragments did not differ between the four conditions of the reading time experiment. The stimuli were presented in booklets in one of four pseudo-randomized orders. The questionnaire began with three examples of continuations. The examples did not contain motion verbs and featured three kinds of prepositional phrases: a locative, a directional telic and a directional ambiguous one. Thirty-two students of the University of Tübingen, all native speakers of German, participated in the study as part of introductory courses in linguistics. They were asked to complete the sentence fragments with locative or directional information that seemed natural to them.  Annotation  The two annotators from the corpus study classified the 1024 completions following the same criteria as in the corpus study. Completions were discarded if they contained adverbials not belonging to the required locative or directional type, particles, absentives, ambiguous completions and completions resulting in a different argument structure. A total number of 659 completions (29 to 61 per verb) remained for the analysis, of which 385 (58%) were telic ones. Each annotator classified half of the completions. 200 randomly chosen items per annotator were cross-checked by the other annotator, yielding 90% agreement. Additionally, two further random samples of 200 items per annotator were cross-checked by two further annotators, yielding 91% agreement.